The study of endogenous cellular factors that inhibit infection by retroviruses is an emerging area of research. We were among the first to show that such factors in monkey cells can inhibit primate lentivirus infection and contribute in a major way to defining retrovirus species tropism. The recent identification of TRIM5alpha as one of this class of retrovirus resistance factors makes this an appropriate time to expand our efforts to characterize these factors. We have obtained preliminary data indicating that TRIM5alpha from humans and nonhuman primates restricts infection by a number of retroviruses, as widely divergent as MLV and HIV-1, and that HIV-1 CA:cyclophilin interaction can modulate restriction factor sensitivity. Other preliminary findings suggest that TRIM5alpha may not be the only capsid-specific retrovirus restriction factor in human cells. This proposal has three specific aims: Aim 1 is to determine the breadth and specificity of retrovirus inhibitory activity exhibited by a number of TRIM5alpha variants and to map determinants of specificity in CA and TRIM5alpha. This will involve the construction of chimeric and mutant human and nonhuman primate TRIM5alpha variants and the selection of HIV-1 mutants that are resistant to TRIM5alpha variants that inhibit wild-type HIV-1. As well as illuminating determinants of restriction specificity in TRIM5alpha and CA, this will provide critical control reagents for Aim 2, which is to determine the mechanism of restriction by TRIM5alpha. We will determine the subcellular localization of TRIM5alpha, attempt to demonstrate physical interaction between HIV-1 capsids and TRIM5alpha, map sequences responsible for TRIM5alpha subcellular localization, multimerization, and capsid binding and determine whether they are required for restriction. We will also determine whether proteins that bind to TRIM5alpha are required for restriction and test the hypothesis that TRIM5alpha-catalyzed ubiquitination is part of the restriction mechanism. Finally, in Aim 3 we will determine whether TRIM proteins that are closely related to TRIM5alpha exhibit retrovirus restriction activity. The successful completion of these studies could have practical implications both for AIDS animal model development, gene therapy vector development and, eventually, for the designing completely novel therapeutic antiretroviral strategies.